NEDA stalls if radical molecular unit is present

NEDA stalls if radical molecular unit is present

Postby Marat.Talipov » Mon Jul 08, 2013 11:56 am

Dear NBO community,

I am trying to run NEDA analysis for a complex [C6F6+* . 2Sb2F11-] at the B3LYPv3/MIDI level, but it stalls in the NBO deletions block.
There was no error message, and the last message was

"""
NEXT STEP: Perform one SCF cycle to evaluate the energy of the new density
matrix constructed from the deleted NBO Fock matrix.

------------------------------------------------------------------------------
"""

The calculation was performed using US Gamess (1 may 2013 R1 revision; gfortran 4.1 + MKL) and NBO6. All the tests coming with the distributives of Gamess and NBO (gmstests, including NEDA) were done successfully. NEDA analysis for another complex (closed-shell complex consisting only of Si, C, and H atoms) also did complete successfully, so I believe the problem is not on a compilation/setup level.

The only unusual thing with the troublesome complex seems to be different alpha/beta electron bonding patterns within the C6F6+* unit, which is caused by the presence of unpaired electron. However, as the alpha and beta molecular units are identical, it should not be a problem for NEDA, at least at the NBO deletions step...

Any thoughts on that?
Attachments
bisallyl.tgz
Gamess Input/Output Files
(188.95 KiB) Downloaded 234 times
Marat.Talipov
 
Posts: 3
Joined: Mon Jul 08, 2013 9:31 am

Re: NEDA stalls if radical molecular unit is present

Postby ericg » Wed Jul 10, 2013 8:11 am

I also encountered problems with this calculation. NEDA evaluates a negative(!) charge transfer energy, then proceeds into the calculation of the fragments. Something is clearly wrong here. I'll let you know what I find.

Eric
ericg
 
Posts: 270
Joined: Sat Dec 29, 2012 9:31 am

Re: NEDA stalls if radical molecular unit is present

Postby Marat.Talipov » Wed Jul 10, 2013 9:52 am

Thanks Eric!

I noticed that NEDA workflow is affected by the DIRSCF option of GAMESS input. With DIRSCF turned off,
calculation complete normally, and with reasonable results. Also, it seems that DIRSCF matters only for DFT calculations,
as the Hartree-Fock NEDA calculations complete normally either with or without DIRSCF.
Marat.Talipov
 
Posts: 3
Joined: Mon Jul 08, 2013 9:31 am

Re: NEDA stalls if radical molecular unit is present

Postby ericg » Wed Jul 10, 2013 9:56 am

Thank you, Marat. This will help me isolate the bug.

Eric
ericg
 
Posts: 270
Joined: Sat Dec 29, 2012 9:31 am


Return to Bug Reports

Who is online

Users browsing this forum: No registered users and 1 guest

cron